VOL. 39, NO. 3 (SEPTEMBER 1986) P. 195-203 ARCTIC

The Controlled Traffic System and Associated in Denali National Park

F.J. SINGER’ and J.B. BEATTIE’

Wildlife Responses

(Received 19 December 1984; accepred in revised form 19 June 1985)

ABSTRACT. Wildlife observability and responses and visitor attitudes were sampled after a mandatory visitor shuttle bus transportation system was in operation for 10 years in Denali National Park, Alaska. Visitor approval of the park transportation policy increased from 80% to 89% from 1972 to 1982-83. Vehicles per day increased 50% from 1973-74 to 1982-83. Moose sightings per trip declined 7296, grizzly bear sightings declined 32%, while Dall sheep and caribou sightings remained constant over the same period. Allowing unlimited private vehicle access during the falls of 1982 and 1983 had little influence upon the numbers of wildlife seen. However, more wildlife were put to flight, flight distances increased, more grizzlies were thwarted from crossing the road, and visitors stopped near and approached wildlife afoot more often. Wildlife responses were significantly influenced by pre-stimulus wildlife behavior for all four species, the type of human behavior for moose and grizzlies, sex/age class for moose and caribou, group size for caribou, number of vehicles present for grizzlies, and the presence or absence of vegetation screening for moose and grizzlies (pc0.05). Moose were alert four times as often (32?21%) when close to the park road than when >1 km away (4.8?5.1%), and when close to the road they were alerted to 37221% of all road stimuli, while caribou were alerted to only 21?11% (p<0.03) of road stimuli. Key words: wildlife viewing, visitor attitudes, wildlife response to traffic, Denali National Park, moose (Alces alces), Dall sheep (Ovis dulli) , grizzly bear (Ursus arcfos), caribou (Rangifer turundus)

RÉSUMÉ. Les possibilités d’observation de la faune, les réactions de la faune et les attitudes des visiteurs furent échantillonnées aprbs la dixibme année d’opération d’un service de navette obligatoire par autobus pour les visiteurs dans le parc Denali en Alaska. Entre 1972 et 1982-83, le pourcentage de visiteurs en faveur de la politique de transport augmenta de 80% à 89%. Le nombre de véhicules par jour augmenta de 50% entre 1973-74 et 1982-83. Les observations d’orignaux par voyage diminubrent de 72%, celles d’ours bruns (grizzlys) baissbrent de 32%, tandis que celles de mouflons de Dall et de caribous demeurtrent constantes durant cette même période. L’accks non contrôllé de vihicules prives permis durant les automnes de 1982 et 1983 produisit peu de différence dans le nombre observations de faune. I1 entraîna cependant la fuite d’un plus grand nombre d’animaux ainsi qu’une augmentation des distances de fuite de ces derniers. Les véhicules empêchtrent aussi àplus d’ours bruns de traverser la route et les visiteurs s ’d ta ien t tout pr& de la faune ou s’en approchaient àpied plus souvent. Les réactions de la faune chez ces quatre espèces étaient influencdes de façon importante par leur comportement habituel pré-stimulus, c’est-à-dire par le type de comportement des humains chez les orignaux et les ours bruns, par classe de sexe et d’âge chez les orignaux et les caribous, par le nombre de véhicules présents chez les ours bruns, et par la présence ou l’absence d’un couvert végetal chez les orignaux et les ours bruns (p<0.05). Les orignaux étaient quatre fois plus vigilants (32221 %) lorsqu’ils se rapprochaient de la route du parc que lorsqu’ilssedéplaçaientà>lkmdecelle-ci(4.8~5.1%).Deplus,lorsqu’ilssetrouvaient~proximitédelaroute,ilsétaienten~tatd’ale~efaceà37~21% des stimulus de la route tandis que caribous n’étaient m i s en état d’alerte que face à 21 ? I l% (p<0.03) de ces mêmes stimulus. Mots cl&: observation de la faune, attitudes des visiteurs, réactions de la faune faceà la circulation, parc national Denali, orignal (Alces alces), mouflon de Dall (Ovis dalli), ours brun (Ursus arcfos), caribou (Rangifer turandus)

Traduit pour le journal par Maurice Guibord.

INTRODUCTION A pervasive challenge in the management of national parks is to balance protection of resources while at the same time making them available for people to enjoy (Everhart, 1972:80-98). At times the two objectives may be contradictory. The manage- ment challenge is nowhere greater than in Denali (formerly Mt. McKinley) National Park, which has become one of the most heavily visited subarctic national parks in the world. Here the opportunity to view the wildlife, tundra ecosystems, and moun- tain scenery, including the highest peak in North America (Mt. McKinley), must be balanced against protectjon of those same sensitive northern resources. We report here on a reevaluation conducted in 1982-83 of both visitor attitudes (Harrison, 1975) and wildlife viewing opportunities (Tracy, 1977) after 10 years of operation of a mandatory shuttle bus system for visitor access along the Denali Park road.

The opportunity to view and conserve highly visible wildlife, particularly moose (Alces ulces), Dall sheep (Ovis dulli), griz- zly bear (Ursus arcros) and caribou (Rangifer turandus), was one of the major reasons that Denali National Park and Preserve was originally created in 1916 (Mt. McKinley National Park Establishment Act 39, Stat. 938) and then greatly expanded in 1980 (Public Law 96-487). In 1936, a 130 km unpavedroad was completed into the park with no restrictions on its use. Visitation

‘National Park Service, 2525 Gambe11 Street, Anchorage, Alaska 99503, U.S.A. @The Arctic Institute of North America

initially was low, since all private vehicles had to be shipped to the park by rail; even after 1957, when the 256 km, arduous, unpaved Denali Highway was built from Paxson to the park, few visitors drove to the park. In 1971, however, the paved George Parks Highway was completed between Anchorage and Fairbanks, making the park accessible to all motorists. Visita- tion subsequently rose from 58 342 in 1971 to 306 027 in 1972. In anticipation of the increase in access, the National Park Service (NPS) in 1973 instituted a mandatory system of free shuttle buses along the park road with the following objectives: 1) to minimize disturbances to wildlife and scenery, 2) to minimize road hazards, and 3) to maximize wildlife and scenery viewing with the least resource impact and energy consumption (U.S. Department of Interior, 1982).

Visitation to Denali Park has greatly expanded since 1972, along with demands on the park road and shuttle bus system. From 1972 to 1982-83, the number of shuttle bus trips per day increased from 27 to 36, the number of other vehicles increased 40% and the number of shuttle bus boardings (X= 2.3 boardings per visitor) increased about 11% annually. The current bus system operates near capacity, with all seats occu- pied on most buses. Increasing the number of buses is a questionable solution because of the quality of the park road base and surface (U.S. Department of Transportation, 1978).

196

STUDY AREA

The Denali Park road traverses the north side of a fault valley that separates the Alaska Range and the Outer Range. This valley is 2-3 km wide near the NPS headquarters at the eastern end, but steeper and only 1 km wide 100 km farther west near Eielson Visitor Center. All wildlife observations were confined to this 100 km section of the park road. The road climbs from 520 m as1 through four passes over 1070 m, travels through one rocky canyon and crosses five braided glacial rivers. Climate is subarctic, with short, cool and rainy summers. Annual precipi- tation is only 38 cm.

Open white spruce (Picea glauca) and tall willow (Salix spp.) stands predominate in the first 50 km of the park road. At higher elevations and along the western half of the road open dwarf shrubs (Dryas spp., Vaccinium spp.) and herbaceous tundra (Carex spp., Eriophorum spp.) predominate (Dean and Heebner 1982).

Dall sheep numbers may have increased from 1978 to 1981, although increased survey efficiency likely accounted for a higher aerial count in 1981 (Singer, 1984). Grizzly bear densi- ties were similar in 1973 and 1983 along the park road (Dean, 1976, 1986). Moose numbers apparently declined 32% in the eastern half of the park road but were stable or more likely increased in the western half 1974-76 to 1984 (Singer and Dalle-Molle, 1985). Caribou numbers increased about 60% since about 1980 (Singer and Dalle-Molle, 1985). Most moose were observed close to the road in the partially forested first 50 km. Dall sheep were usually sighted at greater distances from the road on the steep slopes of the mountain ranges, but sheep were adjacent to the road at Igloo Creek Canyon and Poly- chrome Pass, where the road enters sheep habitat, and also when sheep crossed the park road during seasonal migrations (Murie, 1944). Some caribou, particularly bulls, reside all along the road corridor during the summer, but larger groups of cows and calves migrating to and from calving grounds approached the western half of the road for brief periods (Murie, 1944; Troyer, 1981).

The first 21 km of the park road is paved and open to all vehicles. Beyond that point, beginning on Memorial Day and through Labor Day of each year, all visitors must travel on the shuttle buses. The only exceptions are fee tour buses, NPS administrative and road construction vehicles and private vehi- cles with special permits. Special permits are given to miners operating beyond the far end of the road at Kantishna and to park visitors using one of the five campgrounds located along the road. The next 26 km is improved gravel and is 7.9 m wide. The remaining 53 km of the study road section is only 5.5 m wide and sinuous.

Shuttle buses depart from the park entrance every half hour 0600-2000 h. Visitors may exit or board at any point along the road, assuming a seat is available. Drivers exercise some control near wildlife by moving on if passengers begin to disturb animals and not allowing visitors to exit any closer than 0.8 km from a grizzly bear.

METHODS

Observations of moose, Dall sheep, grizzly bear and caribou were made in 1982 and 1983 from May to early September by driving round-trip from park headquarters to Eielson Visitor Center in an inconspicuous government truck or van. We completed 53 round trips in 1982 and 72 in 1983; Tracy (1977)

F.J. SINGER and J.B. BEATTIE

completed 70 trips in 1973-74 (N= 195 total trips). Vehicle restriction effects were sampled with 16 trips both during four days before and four days after lijting of restrictions on 6 September 1982, 12 trips before and 19 trips after 5 September 1983, and 13 trips before and 13 trips after placement of restrictions on 27 May 1983. During these periods, four differ- ent observers traveled the road each day beginning at 0600, 0800, 1200, and 1400 h respectively. We continued to observe the paved first 21 km section during these before and after tests, since this section was always open to private vehicles and their number also increased here during the open periods.

Each time a group of animals was observed, the vehicle was stopped as close as possible to the animal, the engine turned off and the animal and stimuli observed for 5 min. A 20-40 X spotting scope and 7 x 35 binoculars aided observations. At each observation site we recorded the time, weather, vegetation type (Level I11 of Viereck et al., 1982), slope, elevation and estimated mean dominant canopy height of vegetation. We recorded the number of animals, their sex/age composition, their distance at first response irrespective of angle to the road, their perpendicular distance from the road and whether or not they were partially screened by vegetation. We estimated the angle between the animal and the stimulus at first response and the total distance moved by the animal in a response. Stimuli were the number of buses, private cars or pedestrians and their activity. Pre-stimulus and post-stimulus or response behavior of wildlife included: bed, feed, stand, stand up, brief glance, alert, drift away, walk away, startled, trot away, run away, group bunch up and group reverse direction. Species specific excita- tion or aggressive responses included: stott, excitation jump, head high-spring trot, tail erect and hind leg up-urination alarm for caribou (Pruitt, 1960; Lent, 1966); erect mane, stiff-legged walk and antler threat for'moose (de Vos, 1958; Geist, 1963); and bluff charge, woof and lateral threat for grizzlies (Stonorov and Stokes, 1972). Distances in 1983 were measured with a range-finder (MKS, Ranging Inc.), visual estimates were made in 1982 combined with pacing tests, and these data were compared to the mid-points of Tracy's (1977) distance groupings. Similar data were collected on caribou responses to vehicles from the little-used Stampede Road 1-6 March 1982, during typical winter conditions of about 0.3 m snow depth and -10" to 0°F.

Longer observations were made of caribou and moose groups from inside a parked vehicle to record their alert reactions to a number of different stimuli. Data collection was similar to that for the round trips, but each animal(s) was observed for as long as it was visible and the duration of each behavior was timed to the nearest minute. Moose were more difficult to keep under observation in tall brush, and the observer often left the vehicle on foot to keep the animal in sight. Although efforts were made not to alert the moose to the observer, some such disturbances were unavoidable.

We contacted 400 visitors in both 1982 and 1983: 200 at intermediate rest stops during their shuttle bus trip, 100 car drivers each fall after travel restrictions were lifted, another 100 car drivers in 1983 in mid-summer, while in 1982 another 100 visitors were contacted at the termination of their 200 km round-trip bus ride. Selection of parties was random and only one person in each party was interviewed. We asked the same questions as Harrison (1975) did in 1972, the difference being that he used questionnaire forms, not interviews. Visitors who agreed to the contact (none declined) were asked for: 1) party

WILDLIFE RESPONSE TO TRAFFIC

size, 2) number of days on bus, 3) length of stay in park, 4) residence, 5) sightseeing, backcountry or campground activi- ties, 6) opinion of NPS transportation policy - strongly approve, approve, neutral, disapprove, or strongly disapprove, and 7) opinion of shuttle bus service - excellent, good, fair, poor, very poor. In 1983, we also asked visitors whether they would desire a greater number of shuttle buses. During both falls, we asked car drivers if they had purposely scheduled their trip to avoid travel restrictions. We attempted not to prompt or react to visitor responses.

Only the probability levels are presented for t-test and Chi- square tests. We used the Mann-Whitney U Test (Siegel, 1956) for comparisons involving individual animals, for comparisons to distances that were categorical (Tracy, 1977) or where distance measurements likely increased disproportionately in error, or for comparisons between wildlife numbers likely influenced by groupings. The effects of several variables on wildlife response categories were tested with the nonparametric Kruskall-Wallis 1-way ANOVA (Siegel, 1956), or one- and two-way ANOVA with multiple runs comparing each response category in turn against all the others pooled (Subprogram. ONEWAY, Nie et al . , 1975). Acceptance levels (p) vary based upon our judgment of sample sizes and test power.

RESULTS

Visitor Attitudes

Residency of visitors interviewed was 32% Alaskan, 62% other 49 states, 4% foreign (primarily western Europe), and 2% Canadian. Party size averaged 2.9k 1.7 (X + s.d.), length ofvisitintheparkwas2.3+1.9days(range1-23),andthenum- ber of days riding the bus was 1.3kO.7 (range 0-7). Alaska residents were more common among both car drivers in the uncontrolled fall period and in car drivers commuting to camp- grounds in mid-summer (p<O.001). Twenty-four (24%) of 101 fall car drivers in 1982 had purposely scheduled their trip to avoid road travel restrictions, but this preference was not associated with residency (p=O.75). Nineteen percent of the fall car drivers did not know of the lifting of travel restrictions and would have preferred to ride a shuttle bus. Harrison (1975) also found nearly double the proportion of Alaskans in cars than in buses.

Approval for the NPS mandatory transportation system increased 17% from 1972 to 1982 and further increased 4% in 1983(p<0.001;Table 1). Approvalbycardriversincreasedfrom 1982 to 1983. Alaskans approved less than did other residents (p<0.002). Shuttle bus riders in 1983 approved more than sum- mercardriversandfallcardrivers(p<O.O5;Table l).Thirty-four

TABLE 1. Attitudes of park visitors toward the shuttle bus system in Denali National Park

Approve Neutral Disapprove Year: Group N % N % N % 1972' 251 71 33 9 67 20 1982 287 87* 34 10 10 3 1983 360 91* 14 4 21 5

a) shuttle bus 191 96 5 2 5 2 b) summer car drivers 88 89 3 3 8 8 c) fall car drivers 81 81* 8 8 10 10

*Indicates significant difference from the value directly above it using t-tests

'From Harrison (1975).

"

(p<0.05).

197

percent of the fall car drivers who approved of the transportation policy also preferred to drive their own vehicles. The typical round-trip shuttle bus trip to Eielson Visitor Center takes 7-10 h over a rough road, but apparently fatigue does not alter visitor opinion. There were no differences in opinion between mid- and end-trip interviews (pKO.05).

We pooled the approve-strongly approve and disapprove- strongly disapprove categories, since strongly approve: approve ratios were 2.5:l in 1972, but only 1.2:l in 1982 and 0.6:l in 1983; we suspect survey techniques varied or perhaps distinc- tion between the categories was not meaningful.

Denali Park visitors who provided reasons for their approval of the NPS transportation policy (n=702) strongly identified the system with a reduction in both traffic congestion (n = 205) and environmental impacts (n = 174), such as wildlife protec- tion and reductions in litter and air pollution. Other reasons provided for approval of the system included: enhances viewing of wildlife and scenery (n= 109), improves safety (n=81), more control on visitor activity (n = 51), promotes access and convenience (n = 32), no costs (n = 28), saves wear on personal vehicles (n = 15), and saves energy (n = 7).

Only 35 (5%) of the parties interviewed in 1982-83 disap- proved of the transportation system: 22 complained of losses to personal convenience, 7 gave no reason, 5 complained of hampered wildlife viewing from the bus, 1 found the bus uncomfortable and 1 objected to his loss of personal freedom.

Wildlife Observations

A total of 656 (n = 45 1 groups) moose, 5609,(n = 662 groups) Dall sheep, 897 (n= 556 groups) grizzly bears, and 3801 (n= 762 groups) caribou observations were recorded during four years of data collection in 1973-74 (n = 70 trips; Tracy, 1977), 1982 (n=53 trips), and 1983 (n=72 trips). In March 1982 we recorded an additional 404 (n = 46 groups) caribou observations along the Stampede Road.

Bull ratios for moose in 1982-83 (41 8 : 100 0 ) were similar to 1973-74 (38 8:lOO O ) , as were caribou bull ratios between

Roadside sex/age ratios were apparently representative of the park's population of moose (W. Troyer, pers. comm. 198 1). but caribou bulls were three times as abundant along the road (1 30 8 : 100 Q ) when compared to aerial surveys of the population as a whole (45 8 : l O O 0 ; Troyer, 1981).

Moose sightings declined 72% per road trip 1973-74 to 1982, grizzlies declined 32% (p<0.05), while vehicles per day in- creased 50% (Table 2). Like Tracy (1977), we observed more moose on the first trip of the day (0600 h) in 1982.

Seasonal migrations and seasonal habitat preferences affected wildlife observations. More moose and fewer caribou were observed during fall than summer, more sheep were observed during spring than summer. Grizzly sightings peaked during fall (Mann-Whitney U tests, p<0.05). Weather had no effect on the number of grizzlies seen per trip, but more caribou were seen on clear days and fewer moose, sheep and caribou when it was snowing (Kruskall-Wallis 1-way ANOVA, p<0.05). Dall sheep often fed closer to the road on lower snow-free slopes in spring, and more were observed crossing the road during fall.

Caribou and grizzlies were significantly closer to the road in 1982 than 1973-74 (p<0.05). A similar trend was suggested for moose. The number of moose groups <75 m from the park road increasedfrom46% in 1973-74 to70% in 1982. Grizzlies <75 m

1982-83 (124 8 : l O O 9 ) and 1973-74 (136 8 : l O O 0 , p>0.95).

~~-

198 F.J. SINGER and J.B. BEATTIE

TABLE 2. Moose, sheep, grizzlies and caribou observed per round trip on the Denali Park road during the visitor seasons, 1973-74, 1982, 1983, before and after placement of vehicle restrictions spring 1983, and before and after lifting restrictions fall 1982 and 1983

~ ~

Vehicles Moose Dall sheep Grizzly bear Caribou Period per day' X s.d. x s.d. 1973-742 77 1982 112

5.8 4.7 34.5 19.7 5.9 4.2 33.1 63.6 1.6 0.4*

1983 33.4 6.2 4.0 0.9* 33.5 12.6

121 2.6 2.7* 22.7 29.3 Fall 1982: before 115

4.3 3.4* 43.9 61.6 2.0 2.6 42.5 28.3 4.0 2.1

after 2.4 2.2 46.7 23.9 18.4 9.8

Fall 1983: before 115 5.4 3.1 17.9 40.3

2.1 1.7 6.0 9.5 6.3 3.0 10.0 7.1 after 144 2.2 3.1 9.6 9.4 4.8 2.8*

Spring 1983: before 14.2 10.5

3.8 3.0 40.7 32.2 after

4.0 3.3 103.2 98.6 2.8 2.5 39.0 30.5 2.3 2.8 84.1 48.4

- x s.d. -

x s.d. -

93

*Indicates a difference from the mean directly above using Mann-Whitney U tests, p<O.O5. 'Data not available for all periods. 'From Tracy (1977) for 1973-74.

TABLE 3. Percentage of groups put to flight and flight distances (X + s.d.) for moose, Dall sheep, grizzly bears and caribou from the Denali Park road in 1973-74, 1982 and 1983, and before and after lifting of travel restrictions during fall 1982 and 1983

~ ~~

Moose Dall sheep Grizzly bear Caribou Distance (m) Distance (m)

Period % fleeing TI s.d Distance (m) Distance (m)

% fleeing ?I s.d % fleeing f s.d % fleeing X s.d

1973-74 14 44 63 5 77 111 1982 26*

12 43

111 144 47

14 99 121 O*

1983 8* 28* 37

30 79* 122 17* 21 70 85 31 17

79 73 104 77

52 Fall 1982: before 13 110 100

32 0 18 42 39 30 46 51

35 81 84 after 4* 57 6 14 95 97

Fall 1983: before 54 94* 174* 171

57 121 0 17 59 62 0 42 after 164* 163 24* 49 31 24 96* 104 0

*Indicates a difference in the proportion of groups put to flight in comparison to the value directly above, Chi-square tests, p<O. 10 or difference from the mean distance directly above, Mann-Whitney U tests, p<0.05.

from the park road increased from 25% to 65%, while caribou groups <200 m from the park road increased from 70% to 90%.

Lifting vehicle restrictions on the Denali Park road in the fall had little influence upon numbers of wildlife seen. However, flight distances increased, proportionately more wildlife were put to flight and there were more visitor disturbances (Table 3). Only grizzly bear sightings were reduced (p<0.05; Table 2). More vehicles aggregated near wildlife groups each fall, visitors left their vehicles and approached wildlife more often, more wildlife groups were alerted and put to flight, more caribou

TABLE 4 . Visitor and wildlife behavior compared before and after lifting of vehicle restrictions in Denali National Park, fall of 1982 and 1983 (n= 122 wildlife observations before and 154 after, 15 grizzly crossing attempts before and 17 after, 22 caribou group sightings before and 16 after)

Behaviors Before After Visitors leaving roadside 10 25* Vehicle j a m s 35 104* Visitors threatened by wildlife 2 6 Caribou excitation jumps 0 11* Wildlife alerts 75 131 Wildlife groups fleeing 25 65 * Grizzly road crossings thwarted 2 9* X + s.d. Vehicles near a wildlife group x + s.d. Visitors out of vehicles near wildlife 6.3+12 10.129

*Significant differences pc0.05 using Chi-square and t-tests. Chi-square tests based upon comparisons of total numbers of wildlife observations both before and after or, in the case of grizzly crossings, the number of crossing attempts both before and after were compared to the number thwarted.

- 3.6 t2 6.2*5*

performed excitation jumps and more grizzlies were temporar- ilythwartedfromcrossingtheparkroad(p<O.O5;Table4).Other flight responses, however, were less consistent and suggest annual variations in observers or in wildlife distribution or responses. For example, moose and sheep were more respon- sive during fall in 1983 than 1974-75, but less responsive in 1982.

Denali Park caribou were more sensitive when wintering along the Stampede Road March 1982. The wintering caribou were subjected to 1-4 vehicle passes per day at random intervals in contrast to 200-250 vehicles passing at more regular intervals along the park road. More caribou groups (27%) were put to flight on the Stampede Road (p<0.05), and increases in flight distance approached significance (p<O. lo), but mean distance of caribou groups from the 2 roads did not differ and the number of excitation jumps per (4 per 100 fleeing caribou) was identical.

Wildlife Responses

Larger caribou groups or groups with calves first reacted when closer to the park road (p<0.05; Table 5) . Moose reacted less when screened by vegetation, but grizzlies reacted more when screened by vegetation (p<0.05). Number of vehicles present and grizzly bear reaction distance correlated at pC0.10.

Moose groups within 400 m of the park road alerted (75%) or fled (38%) more than grizzlies (51% alerted, 21% flight) or caribou (51% alerted, 28% flight). Dall sheep were very responsive within 400 m of the road (alert 80%, flight 38%), when they were often far from security habitat and crossing the road. Four running retreats by sheep of 0.4-2.0 km were

WILDLIFE RESPONSE TO TRAFFIC 199

TABLE 5. Regression analysis of factors (Xi) suspected of influencing perpendicular distance (Y) and reaction distance from (Y) from the Denali Park road for responses by moose, Dall sheep, grizzly bears and caribou in 1983; independent variables were number of vehicles, group size, screened:not screened and young of the year present:absent

First reaction distance: Perpendicular No. vehicles Group size Screened Young of year present

No. vehicles Group size Screened Youne of vear Dresent

Perpendicular reaction distance:

Moose (n = 43 groups) Sheep (n= 17) t F f F

0.57 55.54* 0.003 0.19 0.0003 0.01 0.12 5.73* 0.003 0.13

0.001 0.05 O.ooOo5 0.002 0.01 0.44 0.02 0.86

0.90 0.007 0.001 0.005 0.005

0.01 0.003 O.ooo9 0.03

161.33* 0.12 0.01 0.08 0.08

0.17 0.05 0.01 0.56

Grizzly (n = 60) t F

0.84 300.4* 0.02 1.16 0.02 1.05 0.11 7.70* 0.03 1.55

0.05 2.9 0.01 0.71 0.14 9.67* 0.02 1.06

Caribou (n = 99) t F

0.73 269.18* 0.006 0.54 0.08 7.97* 0.005 0.48 0.012 1.27

0.004 0.41 0.08 8.58* 0.005 0.50 0.06 6.27*

* Significant correlation, p<0.025.

observed. Another group remained near the road for three hours attempting to cross. However, when distant sheep sightings were considered they were the least responsive species (14% reacted, 7% fled). Percentage of extreme responses (running, excitation, alarm) was higher for caribou (1 8% of all sightings), followed by moose (14%), grizzlies (10%) and sheep (5%). Caribou also moved farther during a response (1 17a73 m, X +sd), followed by sheep (105295 m), grizzlies (81k79 m) and moose (72+72 m).

Distance from Denali Park road influenced behaviorresponses in all four species (p<O.OOOl; Table 6). Moose, sheep, and caribou responses suggested the following order of increasing intensity of responses with decreasing distance from the road: no response > stare, stand up > walk away > startle, trot, run > excitation.

Group type of caribou significantly influenced behavioral reactions to the road. Using the response ranks above (Kruskall- Wallis one-way ANOVA, p<0.05), the influence of group type on a trot or run response was: cow/calf groups (35%) > mixed (30%) > cows (10%) > bulls (8%). Increasing numbers of vehicles present near a group increased caribou responses (H= 14.9, pc0.05) and grizzly bear responses (H= 39, p<0.05). Human activity influenced moose and grizzly responses, pre-stimulus wildlife behavior affected responses of all four species, while sedage class influenced moose reactions (Table 7). Travelling wildlife were the most likely to respond. Pre-stimulus behavior for grizzlies influenced reactions in the followingorder: walking(71%reacted) > bedded(33%) > stand-

,

ing (12%); the order in 1983 was very similar: walking (79%) > bedding (67%) > feeding(33%). Forcaribou, however, the 1982 pre-stimulus behavior rank was: travelling (77% reacted) > bedding (33%) > standing (12%). Cow moose with calves were more responsive to road stimuli; 80% of cows with calves reacted to the road, 65% of cow only groups, 50% of mixed sex groups, and 48% of bull only groups.

Alert Response of Individual Groups of Moose and Caribou, 1983

Moose alert time was negatively correlated with distance fromtheroad(Spearmanrankcorrelation, r,=-0.53, p<0.05), and caribou alert time was negatively correlated at the p<O. 10 level. Moose were more responsive than caribou and alerted to 37 +21% (X +sd) of all road stimuli (8.6 alerts-h), compared to only 21+- 11% for caribou (3.2 alerts.h, p<0.03). Moose and caribou differed in the stimuli they alerted to (Table 8), with no correlation between their respective rankings of alert stimuli (r, =0.36, p=0.25). Both species alerted highly to the com- bined stimuli of cars and buses, and cars andpedestrians. Moose alerts varied for different stimuli (p<0.03; Table 8), and differ- ences in alerts for caribou approached significance (x2 = 10.1,

For comparative purposes, an additional five moose groups (4 bull, 1 cow groups) were observed for 969 min (1 94+ 108 min per session) in undisturbed situations more than 11 cm from the park road. Moose were alert four times as often (alert 32+2% of

p<O.lO).

TABLE 6 . Perpendicular distance (X + s.d. m) from the Denali Park road for 6 behavioral responses by moose, Dall sheep, grizzly bear and caribou, 1982-83; differences in mean distances between response categories were tested by one-way analysis of variance

Behavioral reswnse

Moose Sheep Grizzly bear caribou N X s.d. N x s.d. N X s.d. N X s.d. - - -

None 128 182 217A 202 570 321A 271 189 205A 217 418 349A Stare, stand up 99 110 106B 116 135 105B 62 76 68B 266 143 lOlB Walk 51 94 193B 13 125 106B 100 120 102B 72 131 l l l B Startle, run, trot 38 55 83B 11 73 69B 39 89 106B 78 104 90B Excitation 7 8 5B 8 98 75B 11 50 68B 44 11 99B Antler, head low, or other threat 1 15 B 2 2.5 3.5B 1 12 B ANOVA Results:

(df = 5) F value 7.6* 65.40* 9.95* 65.96*

A, B -Distances followed by the same letter are not different (pC0.05) according to the least-significant difference procedure. * p<O.OOl for F value.

200 F.J. SINGER and J.B. BEATTIE

TABLE 7. Analysis of variance for behavioral responses of wildlife to the categories of visitor presence, pre-stimulus of wildlife behavior and wildlife sex/age class

Independent variable

Dependent response F P = F P = F P = F P =

Moose Sheep Grizzlies Caribou

Visitor react 1.69 0.19 0.64 0.20 1.32 0.25 0.72 0.54 presence’ alert 0.60 0.52 0.82 0.44 4.94 0.03* 1.46 0.22

walk 2.52 0.08 0.32 0.73 5.73 0.02* 0.29 0.84 alarm 13.25 0.001* 0.19 0.96 1.31 0.27 1.09 0.36

he-stimulus react 0.80 0.54 0.54 0.66 2.72 0.04* 3.59 0.02* behavio? alert 0.61 0.66 1.15 0.35 1.10 0.38 4.37 0.006*

walk 28.84 0.001* 6.48 0.002* 4.84 0.003* 7.87 0.001* alarm -0.002 0.99 0 0.33 0.57 0.02 0.96

Sexlage react 5.14 0.002* 0.15 0.93 1.76 0.18 0.97 0.38 alert 1.86 0.14 0.68 0.57 0.13 0.87 0.17 0.83 walk 0.47 0.71 0.25 0.86 1.58 0.21 0.04 0.95 alarm 2.05 0.11 0.28 0.84 0.50 0.61 0.02 0.75

*Indicates significance at p<0.05. ‘Visitor presence =none, cars and buses but no pedestrians, pedestrians alone or with cars and buses also present. *Pre-stimulus behavior = bed, feed, stand, walk; sexlage = male, female, mixed sex, female with young.

TABLE 8. Alert responses by moose and caribou to various road stimuli during longer observation sessions along the Denali Park road in 1983’

Stimuli Moose Caribou

No. stimuli No. alerts % alerts No. stimuli No. alerts % alerts Cars and buses 13 8 62 57 16 28 Cars and pedestrians 10 6 60 27 8 30 Pedestrians only 28 11 39 13 3 23 Cars only 98 25 26 350 86 25 Buses only 46 12 26 416 72 17 Heavy equipment 17 5 29 Cars, buses, pedestrians 12 1 8 Buses and pedestrians 12 0 0 Other 12 1 8 TOTAL 207 63 904 191

‘Sample size included 35 caribou groups observed for 4762 min (X fs.d. = 217 f 109 min.session), and 23 moose groups observed for 1345 min (120 f 90 min.session). The caribou groups included 14 bull, 4 cow, 4 cow with calf, and 3 mixed groups. The moose groups included 8 cow, 8 cow with calf, 6 bull, and 1 mixed group.

observation time) when close (186t 158 m) to the parkroad (8.6 a1erts.h) than when observed more than 1 km from the road (alerts 4 .8251% of observation time, 2.6 alertsah, p<O.OOOl). Some alerts were to the observers, but this was a consistent factor both close to and distant from the road. We also observed cows with calves to alert to other stimuli such as magpies or other moose.

DISCUSSION

Visitor Attitudes

There were positive biases toward the bus system in our survey methods, but less so than for the 1972 survey (Harrison, 1975). Like Harrison (1973, we did not survey any visitors who had impatiently left the park due to an overcrowded bus and would likely have had more negative responses. We did not mail questionnaire forms to a random sample of Alaska residents, some of whom may not have visited the park because of disapproval of the transportation policy. Harrison (1975) veri- fied that this latter group was more negative about the bus system. However, we eliminated three other possible positive biases of Harrison (1975): 1) the shuttle bus system was no longer in an experimental setting, which might have caused visitors to be less critical of it in 1972; 2) bus drivers distributed questionnaires in 1972 and may have been more helpful and

courteous on those trips; 3) we provided no explanation or justification of the policy, as the questionnaire in 1972 did, which could have introduced a positive bias. The ratio of strongly approve:approve in 1982 was only half that of 1972, and we received many pragmatic statements concerning the policy, such as “the buses are the only way to handle all these people” or “the service was good considering these are just school buses.” This less enthusiastic approval than in 1972 further suggested our survey had less positive bias.

Ratings of bus service and policy approval increased from 1982 to 1983 and fewer complaints were heard. In 1982, we recorded 16 complaints of bus maintenance and 25 of schedule, but zero for either in 1983. Fifty-two percent of bus riders voiced complaints in 1982, but only 28% in 1983. We attribute the decrease in complaints to a change in the bus contractor in 1983 and an improvement in service.

Wildlife Observability

We documented a large degree of variability in the numbers of wildlife observed per trip, in the distances wildlife were observed from the park road and in flight distances. Factors that greatly affect wildlife observability include weather - particu- larly fog, rain and snow, which obscure visibility - diurnal activity fluctuations in wildlife (de Vos, 1958; Roby, 1978) and seasonal migrations of wildlife. We restricted our observations

WILDLIFE RESPONSE TO TRAFFIC

to four days before and after travel restriction changes in order to reduce these influences. Tracy (1977) observed wildlife from shuttle buses, while we primarily used pickup trucks. However, a comparison of 12 bus and 15 pickup truck trips during four days in 1982 suggested no difference in wildlife observability (pC0.05). DistancesweremeasuredbyusandestimatedbyTracy (1977). In spite of these sources of variation, we propose our data reflect major trends in wildlife observability and responses.

Moose and grizzly bears may have been displaced from the park road since 1973-74 or a major change in bear distribution or population size occurred. A grizzly population decline is unlikely (Dean, 1976, 1986), but moose apparently declined about 32% in the east end of the road corridor but not in the west (Singer and Dalle-Moile, 1985). Moose, grizzlies and caribou were observed closer to the road in 1982-83, suggesting habitu- ation by at least some individuals. Van Ballenberghe (1978) observed that certain radio lollared moose exhibited more fear and avoidance of the tranhlaska pipeline than did other collared individuals. Fewer declines in wildlife sightings after the first trip of the day in 1982-83 further suggested habituation. In contrast, ungulate sightings in Elk Island National Park declined by two-thirds after the first passes by Nordic skiers in a day (Ferguson and Keith, 198 l) , and elk in a national park left open meadows after the first vehicle passes in the morning (Schultz and Bailey, 1978).

Wildlife Response Moose: Moose observations declined along the road more

since 1973-74 in comparison to the other three species, although a portion of the drop may have been due to a population decline. They alerted more to stimuli than caribou did, but they responded less by fleeing than the other three species after the lifting of vehicle restrictions. Moose often exhibit a delayed escape response (de Vos, 1958; Altman, 1958; Geist, 1963, 1975). Moose responded less than elk or deer (Odocoileus hemionus and 0. virginianus) and more than bison (Bison bison) and were more likely to remain still or walk away than flee, but were displaced more from ski trails (Ferguson and Keith, 1981).

Moose responded less when screened by vegetation. Simi- larly, elk screened themselves from logging or road building by topographic features and lived within 300 m of a heavily traveled interstate when hidden from passing motorists by trees (Ward e ta l . , 1973).

Alert responses by moose in 1982-83 increased 29% over 1973-74 and flight responses increased 11%, although moose sightings declined. We suspect that some conditioned moose remained close to the park road in 1982-83, but they responded when at similar distances to the road. Caribou moved closer to the Trans-Alaska Pipeline haul road (Roby, 1978) after conditioning to it.

Dall sheep: The greater sensitivity of Dall sheep attempting to cross the road was predicted by higher mean heart rate elevations in bighorns when disturbed on flat terrain or closer than 200 m from a road (MacArthur et al . , 1979).

Sheep showed evidence of habituation in areas where the road passed through steep terrain. Numbers of sheep observed since 1973-74 have remained fairly consistent. Migrating Dall sheep, however, were rarely recorded, and any road crossing avoid- ance could go undetected and might have serious biological consequences (Dean and Tracy, 1979).

Grizzly bears: Grizzlies were observed closer to the road than the other three species, exhibited fewer flight responses and

20 1

were observed closer to the road in 1982-83 than 1973-74. However, grizzlies were the most responsive species in terms of increased flight distances after lifting vehicle restrictions, and fewer bears were observed 1982-83. Grizzly bears are more sought for viewing and therefore subjected to more active human disturbances (Tracy, 1977). Grizzlies apparently felt less secure when screened by vegetation from the road than when they were not, which may be a consequence of poor eyesight and ability to focus on the human shape (Craighead and Craighead, 1966), or a result of greater security when they have the source of human disturbance located (Servheen, 1983).

Caribou: Caribou are considered among the most sensitive of all arctic wildlife species (Geist, 1975; Miller and Gunn, 1979; Klein, 1980), yet our data do not entirely support this conclu- sion. Caribou sightings did not decline 1982-83, and caribou were observed closer to the road. Caribou alerted to the roadway less than moose and showed less variability in their alert responses to different stimuli than moose (range 0-30% versus 8-62%). Unquestionably, the high proportion of resident bull caribou close to the park road contributes to these findings, since in all studies they are less responsive than cow or mixed groups. Also, we sampled many resident, conditioned caribou - flight distances of caribou rapidly migrating through High- way Pass in 1984 were almost two times what we observed along the entire road corridor (J. Beattie, unpubl. data). Flight distances were also slightly longer for Denali caribou along the Stampede Road, where we observed proportionately two-thirds fewer bulls (46 8:lOO 9 ) . Flight distances for caribou were longer in winter in other areas (Bergerud, 1974; Horejsii, 1981) than for the Denali road in summer, so that greater winter responsiveness of caribou (Lent, 1966; McCourt and Horstman, 1974; Shank, 1979) might explain the greater caribou sensitivity along the Stampede Road. We recommend caution when com- paring our caribou response data to other areas because of differences in caribou responses with respect to season and insect harassment (Cameron and Whitten, 1979) and because of the high proportion of bull and resident, conditioned caribou in our sample. However, we know of no differences in sex and age ratios of caribou, seasonal migration patterns or distributions along the road in the years 1973, 1974, 1982 or 1983.

In agreement with other workers, we found caribou groups with calves to be more sensitive to disturbances (Calef et al . , 1976; Miller and Gunn, 1979; Shank, 1979; Klein, 1980), larger groups to be more sensitive (Calef et a l . , 1976; Horejsii, 1981) and traveling caribou to be more sensitive (Surrendi and DeBock, 1?76; Fischer etal. , 1977; Shank, 1979). We observed the order of sensitivity for pre-stimulus behavior for caribou to be travel > bed > stand > feed, but Shank ( 1979) ranked the order travel > feed > bed > stand.

Effects of Lifting Vehicle Restrictions Visitor disturbances to wildlife increased when vehicle restric-

tions were lifted each fall. These brief comparisons represent only partial tests of the lifting of vehicular restrictions, since the number of vehicles increased only 50% but 300-500% increases are predicted if a mid-summer lift of restrictions occurred. Also, four-day test periods were probably too brief for any observable wildlife avoidance to occur. Visitors leaving vehicles accounted for much of the increased wildlife disturbances. Dean and Tracy (1979) reported the park shuttle buses tend to keep people in vehicles, and they observed a doubling in strong reactions in caribou when people got out of a bus. Similarly, elk (Ward et

202

al., 1973; Schultz and Bailey, 1978), bighorn sheep (MacAr- thur et al., 1979) and deer (Kucera, 1976) respond more to skiers or hikers approaching afoot.

MANAGEMENT IMPLICATIONS

Our data indicate the NPS policy of a mandatory public transportation system has been highly successful in Denali National Park relative to its original goals. Visitors’ opinions of the policy is high and has increased from 1972 to 1982-83. Caribou and Dall sheep sightings are nearly identical, and grizzly bear sightings decreased only a minor amount in spite of a 50% increase in vehicles. Evidence points to wildlife habitua- tion since 1973-74. Caribou, grizzlies and, to a lesser extent, moose are seen closer to the park road and flight distances for grizzlies declined. No sexlage ratio changes have been detected. There are fewer reductions in sightings after the first trip of the day. The drop in moose sightings over the eight-year period may be partially explained by a decline in numbers in part of the road corridor and may not have serious ecological implications if moose are only moving short distances from sight into heavy cover or if summer habitat is not limiting. Human activity along roads and trails in other national parks in North America results in much greater wildlife avoidance.

We caution, however, against any expansion in road use. The drop in moose sightings is much larger than any possible decline in numbers. A higher proportion of moose were disturbed by road travel than in 1973-74. Grizzlies were persistently sought after for viewing when close to the road. Dall sheep were highly sensitive during migrations across the road corridor, and any drop in migrational crossings might have been undetected by our data. Denali caribou herd numbers were extremely low, and disturbance to them should be minimized.

Any management measures that reduce the proportion of visitors leaving vehicles will reduce wildlife disturbance rates. Most of the increased wildlife responses after fall lifting of travel restrictions were due to visitors leaving vehicles. More sheep and caribou responses occurred during summer when visitors were out of vehicles than when vehicles alone were present. Regular spacing of vehicles may promote closer approach by wildlife to a road (Roby, 1978; Geist, 1978). However, for species highly sensitive during crossings, such as Dall sheep, or persistently pursued by humans in vehicles, such as grizzlies, clumping of vehicles or caravans may provide longer openings for crossings (Dean and Tracy, 1979).

ACKNOWLEDGEMENTS

The National Park Service, Alaska Region, provided funding. Denali National Park Superintendent R. Cunningham initiated the study and maintained a strong interest in it. NPS employees A. Lovaas, J. Dalle-Molle, J. Cook, R. Contor, R. Peterson, D. Cuillard, R. Kremer, T. Griffiths and B. Swift provided valuable administrative and logistic support, and D. Cuillard additionally provided personnel. K. Laing, S. Duff, K. Kertell, M. Britten, L. Johnson, S . Detweiler, S . McCollough and L. Dalle-Molle worked long hours observing wildlife and in traveling the park road, and D. Miquelle in locating moose. V. Van Ballenberghe, G. Wright, D. Tracy and K. Laing made valuable comments on the study techniques, G. Smoke provided statistical advice, and D. Tracy and F. Dean graciously permitted us to quote from unpublished data.

F.J. SINGER and J.B. BEATTIE

REFERENCES

ALTMANN, M. 1958. The flight distance in free-ranging big game. Journal of Wildlife Management 22:207-209.

BERGERUD, A.T. 1974. Rutting behavior of Newfoundland caribou. In Geist, V., and Walther, F., eds. The Behavior of Ungulates and its Relation to Management. Morges, Switzerland: IUCN Publications New Series No.

BOERTJE, R.D. 1981. Nutritional Ecology of the Denali Caribou Herd. M.Sc. Thesis, University of Alaska, Fairbanks. 294 p.

CALEF, G.W., &BOCK, E.A., and LORTIE, L.J. 1976. The reaction of barren-ground caribou to aircraft. Arctic 29:201-212.

CAMERON, R.D., and WHITTEN, K.R. 1979. Seasonal movements and sexual segregation of caribou determined by aerial survey. Journal of Wildlife Management 43(3):626-633.

CRAIGHEAD, F.C., and CRAIGHEAD, J.J. 1966. Trailing Yellowstone’s grizzlies by radio. National Geographic Magazine 134:252-267.

DEAN, F.C. 1976. Aspects of grizzly bear population ecology in Mount McKinley National Park. In: Pelton, M.R., Lentfer, J. W., and Folk, G.E., eds. Bears - Their Biology and Management. Morges, Switzerland IUCN New SeriesNo. 40. 111-119.

-. 1986. Brown bear density in Denali National Park, Alaska - 1983. Paper presented at the 7th International Conference on Bear Research and Management, in Williamsburg, Virginia, 21-26 February 1986. 30 p. Available from National Park Service, 2525 Gambell Street, Anchorage, Alaska 99503, U.S.A. - and HEEBNER, D.K. 1982. Landsat-based Vegetation Mapping of Mount McKinley National Park region. Final Report for National Park Service Contract No. CX-9OOO-6- E084. Fairbanks: University of Alaska, Alaska Cooperative Park Studies Unit. 198 p.

DEAN, F.C., and TRACY, D.M. 1979. McKinley’s shuttle bus system and the management of traffic impact on wildlife. In: h e r , R., Potter, D.R., Agee, J.K., Anscheu, A., eds. Proc. Recreational Impact on Wildlands. USDA Forest Service, Pacific Northwest Region, No. R-6-001-1979 Seattle. 4-9.

DE VOS, A. 1958. Summer observations on moose behavior in Ontario.

EVERHART, W.C. 1972. The National Park Service. New York: Praeger Publishers, 276 p.

FERGUSON, M.A.D., and KEITH, L.B. 1981. Interactions of Nordic Skiers with Ungulates in Elk Island National Park. Wildlife Technical Bulletin No. 6. Edmonton: Alberta Energy and Natural Resources. 23 p.

FISCHER, C.A., THOMPSON, D.C., WOODEND, R.L., and THOMPSON, P.S. 1977. Ecofogical Studies of Caribou on the Boothia Peninsula and in the District of Keewatin, NWT, in 1976 with Observations on the Reaction of Caribou and Muskoxen to Aircraft Disturbance, 1974-76. Victoria, B.C.: Renewable Resources Consulting Services Ltd.

GEIST, V. 1963. On the behavior of the North American moose (Alces alces andersoni Peterson 1950) in British Columbia. Behavior 20:277-416.

-. 1975. Harassment of large mammals and birds. Report to the Berger Commission. Calgary: Faculty of Environmental Design, University of Calgary. 62 p.

-. 1978. Behavior. In: Schmidt, J.L., and GILBERT, D.L., eds. Big Game of North America. Hanisburg: Stackpole Books. 283-286.

HARRISON, G.S. 1975. The people and the Park reactions to a system of public transportation in Mt. McKinley National Park, Alaska. Journal of Leisure Research 7:6-15.

HOREJSII, B.L. 1981. Behavioral response of barren ground caribou to a moving vehicle. Arctic 34(2):180-185.

KLEIN, D.R. 1980. Reactions of caribou and reindeer to obstructions - a reassessment. In: Reimers, E., Gaare, E., andskjenneberg, S . , eds. Proceed- ings, Second International Reindeedcaribou Symposium. Trondheim, Norway: Direktoratet for Vilt Og Ferskvannsfisk. 519-527.

KUCERA, E. 1976. Deer flushing distance as related to observer’s mode of travel. Wildlife Society Bulletin 4:128-129.

LENT, P.G. 1966. Calving and related social behavior in the barren-ground caribou. Zeitschrift fur Tierpsychologie 6:701-756.

MAcARTHUR, R.A., JOHNSTON, R.H., and GEIST, V. 1979. Factors influencing the heart-rate in free-ranging bighorn sheep: a physiological approach to the study of wildlife harassment. Canadian Journal of Zoology

McCOURT, K.H., and HORSTMAN, L.P. 1974. The reaction of barren ground caribou to aircraft. In: Jakimchuk, R.D., ed. The Reaction of Some Mammals to Aircraft and Compressor Station Noise Disturbance. Renewable Resources Consulting Services Ltd., Canadian Arctic Gas Study Ltd., Biological Report Series 23:l-36.

24. 395-435.

J o ~ a l of Mammalogy 38:128-139.

57:2010-2021.

WILDLIFE RESPONSE TO TRAFFIC

MILLER, F.L., and GUNN, A. 1979. Responses of Peary caribou and Muskoxen to Turbo-helicopter Harassment, Prince of Wales Island, North- west Territories, 1976-77. Edmonton: Canadian Wildlife Service. Occa- sional Papers No. 40. 90 p.

MURIE, A. 1944. The Wolves of Mount McKinley. Fauna Series No. 5. Washington, D.C.: U.S. Government Printing Office. 238 p.

NIE, N.H., HULL, C.H., JENKINS, J.G., STEINBRENNER, K., and BENT, D.H. 1975. Statistical Package for the Social Sciences. New York: McGraw- Hill. 675 p.

PRUITT, W.O., Jr. 1960. Behavior of the Barren-Ground Caribou. University of Alaska, Biological Papers No. 3. 43 p.

ROBY, D.D. 1978. Behavioral Patterns of Barren-Ground Caribou of the Central Arctic Herd adjacent to the Trans-Alaska Pipeline. M.Sc. Thesis, University of Alaska. 200 p.

SCHULTZ, R.D., and BAILEY, J.A. 1978. Responses of National Parkelk to human activity. Journal Wildlife Management 42:91-100.

SERVHEEN, C. 1983. Grizzly bear food habits, movements, and habitat selecting in the Mission Mountains, Montana. Journal of Wildlife Manage- ment 47(4):1026-1035.

SHANK, C.C. 1979. Human-Related Behavioral Disturbance to Northern Large Mammals: a Bibliography and Review. Report prepared for Foothills Pipe Lines (South Yukon) Ltd., Calgary. 254 p.

SIEGEL, S. 1956. Nonparametric Statistics for the Behavioral Sciences. New York: McGraw-Hill. 312 p.

SINGER, F.J. 1984. Some Population Characteristics of Dall Sheep in Six National Parks and Preserves. In: Hoefs, M., ed. Proceedings, Fourth Northern Wild Sheep and Goat Conference. Whitehorse, Yukon: Game Branch. 1-10. - and DALLE-MOLLE, J. 1986. Monitoring the Denali Moose-

Caribou-Predator System. Alces 21:339-358.

203

STONOROV, D., and STOKES, A.W. 1972. Social behavior of the Alaska brown bear. In: Herrero, S., ed. Bears - their biology and management. Morges, Switzerland IUCN Publications, New Series No. 23. 232-242.

SURRENDI, D.C., and DEBOCK, E.A. 1976. Seasonal Distribution, Popula- tion Status and Behavior of the Porcupine Caribou Herd. Canadian Wildlife Services Report to MacKenzie Valley Pipeline Investigation, Ottawa. 144 p.

TRACY, D. 1977. Reactions of Wildlife to Human Activity along Mount McKinley National Park Road. M.Sc. Thesis, University of Alaska, Fair- banks. 260 p.

TROYER, W. 1981. Movementsofthe Denali CaribouHerd. Anchorage: U S . National Park Service. Unpubl. report. 19 p. Available from Alaska Regional

99501, U.S.A. Office, National Park Service, 2525 Gamhell Street, Anchorage, Alaska

U.S. DEPARTMENT OF INTERIOR, National Park Service. 1982. Environ- mental Assessment for the Development Concept Plan, Denali Park Road Corridor. Denver Service Center, National Park Service, U.S. Department of Interior, Denver, Colorado. 30 p.

U.S. DEPARTMENT OF TRANSPORTATION, Federal Highway Adminis- tration. 1978. McKinley Park Highway Study Report, Office of Federal

VAN BALLENBERGHE, V. 1978. Final Report on the Effects of the Trans- Highway Projects, Vancouver, Washington. 35 p.

Alaska Pipeline on Moose Movements. Special Report No. 23, Joint State/ Federal Fish and Wildlife Advisory Team, Anchorage, Alaska. 41 p.

VIERECK, L., DYRNESS, C.T., andBATTEN, A.R. 1982.1982Revisionof Preliminary Classification for Vegetation of Alaska. Institute of Northern Forestry, U.S. Forest Service, 300 Tanana Drive, Fairbanks, Alaska. 72 p.

WARD, A.L., CUPAL, J.J. , LEA, A.L., OAKLEY, C.A., and WEEK, R.W. 1973. Elk Behavior in Relation to Cattle Grazing, Forest Recreation, and Traffk. Transactions North American Wildlife Conference 38:327-337.